Building Moon Bot: A Slack-Native Coding Agent Backed by HuggingFace Buckets
โข 53
AI & ML interests
The AI community building the future.
Recent Activity
View all activity
Papers
Can We Defend Against AI-Generated Video Attacks on Real-World Crisis Events? A Systematic Evaluation of Detectors, Generators and Social Dissemination
Seeing the Needle in the Haystack: Towards Weakly-Supervised Log Instance Anomaly Localization via Counterfactual Perturbation
Articles
nielsrย
updated a
bucket about 3 hours ago
tarekziadeย
updated a
bucket about 6 hours ago
alvarobarttย
updated a
dataset about 6 hours ago
Add `microsoft-foundry` assets for the documentation
#661 opened about 15 hours ago
by 
alvarobartt
celinahย
updated a
bucket about 19 hours ago
sayakpaulย
updated a
dataset about 24 hours ago
aminediroHFย
updated a
dataset 1 day ago
sergiopaniegoย
posted an update 8 days ago
Post
2006
Can you do RL over taste?
I've spent some time reproducing, in the open, Surya N's idea of training a model to paint with code. It's a coding model that learns to paint watercolours by writing JS code, trained with GRPO. I used TRL and OpenEnv for this, with the whole pipeline running on Hugging Face.
The interesting part is that the reward has no correct answer, unlike a math problem. In this case it's based on the artistic preferences of the person who builds the dataset.
Everything is published: the environment, the reference pool, the trained adapters, every painting of every run with the code that made it, and a write-up with all the decisions, including the ones that went wrong.
Blog post: https://huggingface.co/blog/train-to-paint-with-code
I've spent some time reproducing, in the open, Surya N's idea of training a model to paint with code. It's a coding model that learns to paint watercolours by writing JS code, trained with GRPO. I used TRL and OpenEnv for this, with the whole pipeline running on Hugging Face.
The interesting part is that the reward has no correct answer, unlike a math problem. In this case it's based on the artistic preferences of the person who builds the dataset.
Everything is published: the environment, the reference pool, the trained adapters, every painting of every run with the code that made it, and a write-up with all the decisions, including the ones that went wrong.
Blog post: https://huggingface.co/blog/train-to-paint-with-code
sergiopaniegoย
posted an update 14 days ago
Post
436
catching up on some bookmarked reads from the summer, reading Antidoom from @liquidai
small reasoning models get stuck more easily when the task involves a long thinking trace and a hard problem. It starts repeating the same word over and over again ("Wait", "Alternatively"โฆ), each repetition makes the next one likelier, and the generation is spent before it reaches an answer
they measured it, 10.2% of completions for an early LFM2.5-2.6B checkpoint and 22.9% for Qwen3.5-4B at greedy. After training those drop to 1.4% and 1.0%
the fix is FTPO (final token preference optimization). What I like is how narrow it is, it only touches the single token where the loop starts
three ways it differs from DPO:
> trains one token position, mid-generation, instead of whole sequences
> spreads probability across ~20 plausible alternatives instead of swapping one overtrained token for another
> keeps the regularizer in logit space, no softmax, so the rest of the vocabulary stays put
the third one is what makes it usable. If you want to edit one position without disturbing the model, you can't have a loss that reshuffles the other 150k logits on the way
and their explanation abt the result: the training teaches the model nothing new about math or code, it clears the failure mode that was blocking answers the model could already produce
full blog > https://www.liquid.ai/blog/antidoom
FTPO itself comes from Antislop, where it was built to strip overused phrasing. LiquidAI retargeted it to doom loops
and under the hood it's a subclass of TRL's DPOTrainer with compute_loss overridden, around 90 lines of loss and no new trainer
we documented that pattern in TRL's docs
https://huggingface.co/docs/trl/main/en/customization#change-the-training-objective
small reasoning models get stuck more easily when the task involves a long thinking trace and a hard problem. It starts repeating the same word over and over again ("Wait", "Alternatively"โฆ), each repetition makes the next one likelier, and the generation is spent before it reaches an answer
they measured it, 10.2% of completions for an early LFM2.5-2.6B checkpoint and 22.9% for Qwen3.5-4B at greedy. After training those drop to 1.4% and 1.0%
the fix is FTPO (final token preference optimization). What I like is how narrow it is, it only touches the single token where the loop starts
three ways it differs from DPO:
> trains one token position, mid-generation, instead of whole sequences
> spreads probability across ~20 plausible alternatives instead of swapping one overtrained token for another
> keeps the regularizer in logit space, no softmax, so the rest of the vocabulary stays put
the third one is what makes it usable. If you want to edit one position without disturbing the model, you can't have a loss that reshuffles the other 150k logits on the way
and their explanation abt the result: the training teaches the model nothing new about math or code, it clears the failure mode that was blocking answers the model could already produce
full blog > https://www.liquid.ai/blog/antidoom
FTPO itself comes from Antislop, where it was built to strip overused phrasing. LiquidAI retargeted it to doom loops
and under the hood it's a subclass of TRL's DPOTrainer with compute_loss overridden, around 90 lines of loss and no new trainer
we documented that pattern in TRL's docs
https://huggingface.co/docs/trl/main/en/customization#change-the-training-objective
- 1 reply
sayakpaulย
authored a
paper 14 days ago
sayakpaulย
submitted a
paper to Daily Papers 14 days ago
eliebakย
authored 2
papers 17 days ago
bezzamย
submitted a
paper to Daily Papers 21 days ago
sergiopaniegoย
posted an update 22 days ago
Post
300
super interesting new paper from Microsoft "Agent Lightning v1.0: Towards Harnessed Agentic RL" by Zhiyuan He et al.
same idea we've seen already several times: you train the agent inside the real harness it ships with, instead of a reimplementation of it
now that recipe has a name โ harnessed agentic RL
paper: huggingface.co/papers/2608.17528
the tricky bit they nail down: one rollout is not one training sample
the harness calls the model many times, so a single episode โ a variable number of (prompt, response) rows
you don't even know the batch size until the episode finishes running
its real contribution is being first to systematically map the four problems that fall out of that:
> retokenization + sample merging
> advantage calculation over a variable sample count
> loss normalization at the rollout level, not per sample
> backend scheduling when the batch size is dynamic
and it actually works โ plain RL inside the real harness, no reimplementation
Qwen3.5-9B on SWE-bench Verified 41.8 โ 56.4 (+14.6), with only ~6k examples
the whole thing is ~3,500 lines, any harness, self-hosted k8s
from our side, we've shared some materials on the same line you may want to check out :)
> Agentic RL: Token-In, Token-Out Done Right: https://huggingface.co/blog/huggingface/tito
> a full worked example, opencode owning its loop trained with GRPO: https://huggingface.co/blog/sergiopaniego/trl-openenv-harness-training
> Harness, Scaffold, and the AI Agent Terms Worth Getting Right: https://huggingface.co/blog/agent-glossary
on a similar line:
https://x.com/SergioPaniego/status/2062911580564496576
same idea we've seen already several times: you train the agent inside the real harness it ships with, instead of a reimplementation of it
now that recipe has a name โ harnessed agentic RL
paper: huggingface.co/papers/2608.17528
the tricky bit they nail down: one rollout is not one training sample
the harness calls the model many times, so a single episode โ a variable number of (prompt, response) rows
you don't even know the batch size until the episode finishes running
its real contribution is being first to systematically map the four problems that fall out of that:
> retokenization + sample merging
> advantage calculation over a variable sample count
> loss normalization at the rollout level, not per sample
> backend scheduling when the batch size is dynamic
and it actually works โ plain RL inside the real harness, no reimplementation
Qwen3.5-9B on SWE-bench Verified 41.8 โ 56.4 (+14.6), with only ~6k examples
the whole thing is ~3,500 lines, any harness, self-hosted k8s
from our side, we've shared some materials on the same line you may want to check out :)
> Agentic RL: Token-In, Token-Out Done Right: https://huggingface.co/blog/huggingface/tito
> a full worked example, opencode owning its loop trained with GRPO: https://huggingface.co/blog/sergiopaniego/trl-openenv-harness-training
> Harness, Scaffold, and the AI Agent Terms Worth Getting Right: https://huggingface.co/blog/agent-glossary
on a similar line:
https://x.com/SergioPaniego/status/2062911580564496576
sergiopaniegoย
posted an update about 1 month ago
Post
743
Something I really like when I study a subject is understanding its history, how it reached the point where it is today
I did that exercise for RL in post-training: from RLHF and PPO, to verifiable rewards, to the GRPO family of variants, to agents acting in environments. Everything is backed by what the labs themselves say in their public reports (DeepSeek, Qwen, Kimi, GLM-5, Nemotron, Mistral and more), in their own words
This is the companion piece to Class 3 of our Training Agents series with @burtenshaw . The class explains how GRPO works, with three hands-on experiments. The article shows where the same ideas appear at frontier scale
https://huggingface.co/blog/sergiopaniego/agentic-rl-2026
I did that exercise for RL in post-training: from RLHF and PPO, to verifiable rewards, to the GRPO family of variants, to agents acting in environments. Everything is backed by what the labs themselves say in their public reports (DeepSeek, Qwen, Kimi, GLM-5, Nemotron, Mistral and more), in their own words
This is the companion piece to Class 3 of our Training Agents series with @burtenshaw . The class explains how GRPO works, with three hands-on experiments. The article shows where the same ideas appear at frontier scale
https://huggingface.co/blog/sergiopaniego/agentic-rl-2026
- 1 reply
sergiopaniegoย
posted an update about 1 month ago
Post
329
we just released a new blog "Training a coding agent using the OpenCode harness in remote HF sandboxes with TRL and OpenEnv"
you can take a real coding agent (OpenCode), let it run its own tool loop against real coding problems, and train it with RL on the exact tokens it produced
and every rollout runs in its own remote HF sandbox, so rollouts scale out beyond one machine
the loop:
- OpenCode owns its tool loop inside an OpenEnv sandbox
- an in-sandbox proxy records the real token ids + logprobs, per turn
- a hidden-test verifier scores the result, and that is the reward
- TRL trains with AsyncGRPO, weights sync back to vLLM over NCCL
blog + runnable example: https://huggingface.co/blog/sergiopaniego/trl-openenv-harness-training
you can take a real coding agent (OpenCode), let it run its own tool loop against real coding problems, and train it with RL on the exact tokens it produced
and every rollout runs in its own remote HF sandbox, so rollouts scale out beyond one machine
the loop:
- OpenCode owns its tool loop inside an OpenEnv sandbox
- an in-sandbox proxy records the real token ids + logprobs, per turn
- a hidden-test verifier scores the result, and that is the reward
- TRL trains with AsyncGRPO, weights sync back to vLLM over NCCL
blog + runnable example: https://huggingface.co/blog/sergiopaniego/trl-openenv-harness-training
- 3 replies
sergiopaniegoย
posted an update about 1 month ago
Post
2658
LFM2.5-2.6B just dropped!
and the @liquidai blog comes with some nice details about the training procedure, so let's analyze it.
basically, a full agent training pipeline but compressed into 2.6B
base model โ SFT โ specialized teachers per domain (SFT + RLVR) โ on-policy distillation back into one student โ agentic RL
the two most interesting stages
โ MOPD: the student generates, each prompt routes to its domain teacher for token-level feedback. teachers branch from the same SFT checkpoint, so their signal stays close to the student's distribution
โ agentic RL: multi-turn GRPO inside real harnesses (OpenClaw, Hermes Agent), one sandbox per rollout, a proxy captures token-level trajectories while the harness stays a black box
this makes a 2.6B that beats much larger models on instruction following and tool use
SFT, distillation, RL, RL envs: exactly what we're covering in our Training Agents livestream series (next one coming soon!)
โ model: LiquidAI/LFM2.5-2.6B
โ blog: https://www.liquid.ai/blog/lfm2-5-2-6b
โ live series: https://www.youtube.com/playlist?list=PLo2EIpI_JMQvQZm-kVlz4wY1vWF0LBcf5
and the @liquidai blog comes with some nice details about the training procedure, so let's analyze it.
basically, a full agent training pipeline but compressed into 2.6B
base model โ SFT โ specialized teachers per domain (SFT + RLVR) โ on-policy distillation back into one student โ agentic RL
the two most interesting stages
โ MOPD: the student generates, each prompt routes to its domain teacher for token-level feedback. teachers branch from the same SFT checkpoint, so their signal stays close to the student's distribution
โ agentic RL: multi-turn GRPO inside real harnesses (OpenClaw, Hermes Agent), one sandbox per rollout, a proxy captures token-level trajectories while the harness stays a black box
this makes a 2.6B that beats much larger models on instruction following and tool use
SFT, distillation, RL, RL envs: exactly what we're covering in our Training Agents livestream series (next one coming soon!)
โ model: LiquidAI/LFM2.5-2.6B
โ blog: https://www.liquid.ai/blog/lfm2-5-2-6b
โ live series: https://www.youtube.com/playlist?list=PLo2EIpI_JMQvQZm-kVlz4wY1vWF0LBcf5
- 3 replies
sergiopaniegoย
posted an update about 1 month ago
Post
2680
Simon Willison (@simonw ) has asked every new model to draw a pelican riding a bicycle for some time now
you look at the drawing and you know. but there is no number, so nothing can train against it, no?
I turned this idea into an rl env in OpenEnv. now, you can eval any model against it, and train against it with TRL
read the details!๐ค
https://huggingface.co/blog/sergiopaniego/pelican-env-openenv
you look at the drawing and you know. but there is no number, so nothing can train against it, no?
I turned this idea into an rl env in OpenEnv. now, you can eval any model against it, and train against it with TRL
read the details!๐ค
https://huggingface.co/blog/sergiopaniego/pelican-env-openenv
- 2 replies
sergiopaniegoย
posted an update about 1 month ago
Post
248
yesterday we had Class 3 of the Training Agents live series
we went through how GRPO works in depth and applied it to real experiments with TRL
sharing the resources in case you want to dig in, enjoy!
๐ฅ session recording: https://www.youtube.com/watch?v=ztdTed5egrM
๐ slides with links: https://docs.google.com/presentation/d/19v5_HR5B-1CPHuoZ-RXjhgBrGFXGnBNd6TfLhsXWT1c/edit?usp=sharing
we went through how GRPO works in depth and applied it to real experiments with TRL
sharing the resources in case you want to dig in, enjoy!
๐ฅ session recording: https://www.youtube.com/watch?v=ztdTed5egrM
๐ slides with links: https://docs.google.com/presentation/d/19v5_HR5B-1CPHuoZ-RXjhgBrGFXGnBNd6TfLhsXWT1c/edit?usp=sharing